Prediction method of Kawasaki disease C-ball non-response, storage medium and terminal equipment

Abstract

The invention discloses a prediction method of Kawasaki disease C-ball non-response, a storage medium and terminal equipment, and the method comprises the steps: carrying out the feature screening of a Kawasaki disease clinical training data set through Lasso, selecting 14 features related to the IVIGRKD; performing a Logistic regression on the 14 features to build a Lasso-Logistic regression model; drawing a column line diagram according to the Lasso-Logistic regression model, marking a scale on a corresponding line segment of each feature to represent a value range of the feature; assigning a score to each feature at each value level, calculating a total score of the 14 features, deriving a Kawasaki disease C-ball non-response prediction result from a function conversion relationship between the total score and an IVIGKD occurrence probability. The invention uses the constructed Lasso-Logistic regression model for predicting IVIGRKD with higher sensitivity and specificity, and draws a visual column line diagram according to the Lasso-Logistic regression model in order to make the prediction model be easier to be understood; besides, clinicians can score in the column line diagram according to the clinical data of patients to obtain a IVIGRKD occurrence probability, so that the method is clinically applicable.

Description

technical field [0001] The invention relates to the technical field of prediction of gamma globulin unresponse in children with Kawasaki disease, in particular to a prediction method, storage medium and terminal equipment for Kawasaki disease gamma globulin unresponse. Background technique [0002] Kawasaki disease (Kawasaki disease, KD), also known as mucocutaneous lymph node syndrome, was first reported by Japanese scholar Tomiyu Kawasaki in 1967. KD mainly occurs in children under the age of 5. The etiology is unknown at present. It is an acute, self-limited febrile rash vasculitis disease, which can involve small and medium blood vessels in the whole body. It has replaced rheumatic fever as acquired heart disease in children in developed countries. It is the number one cause of disease, and its incidence rate shows an increasing trend year by year. The etiology and pathogenesis of Kawasaki disease have not been fully clarified, and inflammatory response is its main path...

AI Technical Summary

Problems solved by technology

[0005] In view of the above-mentioned deficiencies in the prior art, the purpose of the present invention is to provide a Kawasaki disease gamma non-response prediction method, storage medium and terminal equipment, aiming to solve the problem that the scoring items of the existing IVIGRKD scoring system are quite different, and the prediction efficiency Sensitivity and specificity vary, showing problems with regional specificity of different scoring systems

Images